The present invention relates to a technique of modifying facial images and a makeup art of putting makeup on a face using the modifying technique.
The present inventor has been making studies, for a long time, on factors which determine the faces of persons, particularly facial impressions, and beauty and ugliness, from the viewpoints of plastic surgery and cosmetic surgery. The study has provided several new findings. One of them is the xe2x80x9chighlight theoryxe2x80x9d. Conventionally, it was a general thought that the components of a face, such as eyebrows, eyes, a nose and lips, would become significant factors to determine a facial impression. The research done by the present inventor however shows that the prime factor to determine a facial impression is the shading state of a face, i.e., the states of a bright area and a dark area, particularly, the state of the brightest area or highlighted area. The shading state relates to shading on a face which occurs under illumination of the natural light, and reflects or expresses the three-dimensional shape of the face. The states of a highlighted area in such shading particularly affects a facial impression.
The analysis based on this highlight theory classifies faces into three general types. As exemplarily shown in FIG. 5, the shapes of highlighted areas are an inverted triangle type [(a) in FIG. 5], an oval type [(b) in FIG. 5] and a clover-leaf type [(c) in FIG. 5]. A typical beautiful face in each facial type has a good shape of the associated highlighted area. With regard to an inverted triangle type, for example, no or little distortion or deformation on its shape is a significant factor in giving a good impression on a face.
Another finding is a theory on the shapes of faces people feel beautiful or preferable. Conventional theories on facial beauty include canon theory or a theory based on golden cut, and the average theory which has recently come up. The average theory means a theory that the face that appears most beautiful among those of a certain human race is an average face of the individual faces of that race. Faces based on those theories merely provide general xe2x80x9cbeautiful facesxe2x80x9d. In other words, those theories cannot give ideal post-makeup faces which are makeup aims applicable to makeup methods. Unlike those theories, the theory found by the present inventor can be put to a practical use in makeup methods.
According to what has been newly found by the present inventor, through the following procedures, it is possible in most occasions to find the most beautiful or preferable face for each specific person, or the most beautiful or preferable face that can be made by makeup available to that individual. The fundamental theory is such that if faces are produced by combining a face of given standards, specifically, a face generally considered beautiful or a face preferred by an individual and the actual face of that particular individual at the proper mixing ratios through an image processing technique, an xe2x80x9cideal facexe2x80x9d which is considered most beautiful or preferable by that individual can be found in most cases from the foregoing mixed faces. Specifically, one of model faces, which are, for example, faces of preferred talents or actresses or actors, or xe2x80x9cbeautiful facesxe2x80x9d originated from the average theory and the like, is selected as a desirable face. Based on this desirable face, the actual face of the person is subjected to image processing like morphing so that the actual face is mixed with the desirable face to approach the latter. This process provides a plurality of xe2x80x9cmodified facesxe2x80x9d with different mixing ratios or similarity levels with respect to the desirable face. In most cases, those xe2x80x9cmodified facesxe2x80x9d contain an xe2x80x9cideal facexe2x80x9d that person considers most beautiful or preferable. The data obtained by the present inventor shows that persons in 48 cases out of 50 could find xe2x80x9cideal facesxe2x80x9d they thought most beautiful or preferable. In the other two cases, xe2x80x9cideal facesxe2x80x9d could be found by modifying the desirable faces. The mixing ratio in every case ranged from 10 to 50%.
It is to be noted that one model face itself should not necessarily be the most preferable face for a particular person. That is, the xe2x80x9cideal facexe2x80x9d a particular person can find for herself or himself is a xe2x80x9cmodified facexe2x80x9d which is resulted from the mixing of a model face with her or his own face, not the model face itself. The actual results of the experiments conducted by the present inventor show that most subject persons did not prefer model faces using xe2x80x9cbeautiful facesxe2x80x9d created by the average theory or the like.
The present invention has been devised to effectively utilize the above-described findings and alms at using those findings to a technique of modifying, for example, photographed facial images and to a makeup technique.
According to the present invention, facial images picked up by a still camera, a video camera or the like are modified as follows. First, a desirable face is determined, and then an original facial image is subjected to image processing to become closer to this desirable face so that a modified facial image is acquired.
Model faces can be used in determining a desirable face. The faces of preferable talents or actresses or actors can be used as model faces. If one does not have any preferable talent or the like, standard good-looking faces acquired from a plurality of beautiful faces which have previously been prepared corresponding to human races or like according to the foregoing average theory may be used. In the foregoing method, it may also be arranged that an original facial image is subjected to image processing to become closer to a desirable face, thereby forming images of a plurality of selectable modified faces with different similarity levels with respect to the desirable face, and then a selection is made from those selectable modified faces to acquire a modified facial image. This further facilitates acquiring a face ideal to the subject person.
It is preferable to use the foregoing highlight theory in modifying and making an original facial image closer to a desirable face. That is, the key point is to make the states of highlighted areas of a facial image to be modified closer to those of a desirable face. This can provide a more beautiful and preferable face and thus facilitate approaching to a desirable face.
This facial image modifying method is an application of the foregoing theory on the shapes of faces that people think beautiful or preferable. Unlike the conventional photograph modification techniques which mainly depend on a manual work, this method can automate the modification work by electronic processing. What is more, modification results can give greater satisfaction. This method of the present invention can be adapted to operational simulation in plastic surgery and cosmetic surgery as well as ordinary photograph modification.
In a makeup simulation method embodying the present invention and a makeup method which uses this simulation method, a desirable face is determined first. In the case where a makeup instructor instructs a subject person how to put makeup, for example, a desirable face may be determined by asking the subject person about preferences. Alternatively, model faces may be used in determining a desirable face. The faces of talents, or actresses or actors the subject person like can be used as model faces. If the subject person has no preference for any talent or the like, averages of good-looking faces which have previously been prepared corresponding to human races or the like may be used as model faces. In the former case, a preferred desirable face can be selected from model faces using the name or the like of the preferable talent or the like as an index. In the latter case, a preferred desirable face should be selected from model faces of standard beauty which has previously been input as data. In both cases a selection is made while sequentially displaying the model faces on a monitor screen or based on the type of the face of the subject person. In the latter case, the model faces should not necessarily be shown to the subject person.
Once the desirable face is decided, then images of imagined post-makeup faces based on the desirable face or post-makeup faces which are imagined as preferable are prepared. This involves a process of mixing the face of the subject person with the desirable face to make the former face closer to the latter by performing image processing like shape-merging on the image of the face of the subject person. Then, an ideal post-makeup face the subject person desires is determined from the imagined post-makeup faces. Specifically, a plurality of imagined post-makeup faces with different mixing ratios or similarity levels with respect to the desirable face are acquired through the foregoing image processing, and a preferable selection is made from those faces as an ideal post-makeup face within the range of the similarity levels which can be obtained by makeup. Thus, the ideal post-makeup face which is expected to be finally obtained is given in advance. That is, the subject person can know the makeup-finished face in a short time. As apparent from the above, the significant feature of the makeup method embodying the present Invention lies in that a final, makeup-finished face can be presented in a short time by determining a desirable face on which a makeup procedure is based and acquiring an ideal post-makeup face with respect to this desirable face. In this feature is effectively used the above-described theory on the shapes of faces people think beautiful or preferable.
When an ideal post-makeup face is decided through the foregoing makeup simulation, a makeup procedure is derived from this ideal post-makeup face. Specifically, a sequence of makeup procedures for achieving the ideal post-makeup face, such as selecting which portions of eyebrows to be shaved and where to draw eyebrows, selecting the lines and ranges for eye liners and eye shadows, selecting the colors of eye shadows, setting the drawing range for a lipstick and separately applying different foundations, is derived from a previously set makeup program. Based on this makeup procedure, makeup is applied to the subject person. This makes it possible to accurately realize the ideal post-makeup face or the makeup state the subject person has previously found appropriate, on the face of the subject person. That is, makeup which the subject person desires can be done freely and in a short time.
To realize the foregoing feature of the makeup method of the present invention which lies in the preparation of an ideal post-makeup face through image processing based on a desirable face or a model face, it is important to mix the current face of a subject person with a model face through image processing to get the former face closer to the latter. That is, it is essential to acquire an ideal post-makeup face by making the current face of the subject person closer to the reference model face by image processing. The feature also requires an effective scheme or standards for making the current face of the subject person closer to the ideal post-makeup face by actual makeup. This can be accomplished by the foregoing highlight theory.
Based on the highlight theory which teaches highlighted areas considerably affect the impression of a face and a beautiful and preferable face can be made by erasing any unnecessary shade (dark area) to make the shapes of the highlighted areas better, it becomes relatively easy to get the face of a subject person closer to a model face through image processing and it is also easy to make the face of the subject person closer to a desirable face by implementing such a makeup as to adjust the states of the highlighted areas in association with those of the model face (desirable face). In other words, a more effective scheme of making the face of the subject person closer to the reference model face can be acquired based on the highlight theory.
The image processing for preparing the images of imagined post-makeup faces on the basis of the highlight theory involves a process of making the shapes of the eyebrows, the eye lines, the lip line, and so forth closer to those of a desirable face within a range where modification by makeup is possible, in addition to the basic process of getting the states of highlighted areas on a face closer to those of the desirable face within a range where modification by makeup is possible. A shape-merging scheme or the like can be also used in this case.
Based on the highlight theory, attention should be paid to a greatest highlighted area as an impressing element of a face in the makeup simulation method and makeup method, too. Since makeup about highlighted areas is accomplished by separately applying different foundations, the types of foundations and procedures for applying them separately are derived as essential factors from an ideal post-makeup face.
To accomplish the process, image processing is performed to divide the ideal post-makeup face into a plurality of areas originating from differences between bright and dark areas like a pattern of contour lines. Then, color data is acquired for each segmented area. Color data in this case consists of hue and brightness, and chroma if necessary, and R, G and B which are used in a television system or C, M, Y and K which are basic printing colors in a computer system are normally used. Based on the color data of the individual areas and a previously set foundation program, foundations to be applied on the individual areas are determined. Finally, the thud determined foundations are applied on the face of the subject person in accordance with the segmented areas.
One makeup simulation method according to the present invention and a makeup method using this simulation method therefore include the following steps: (a) determining a desirable face and performing image processing on an image of a face of a subject person based on the desirable face, thereby forming images of imagined post-makeup faces; (b) displaying the imagined post-makeup faces, prepared in the step (a), on a monitor screen or the like and selectively determining an ideal post-makeup face desired by the subject person from the displayed imagined post-makeup faces; (c) deriving a makeup procedure from an image of the ideal post-makeup face, determined in the step (b), based on a previously set makeup program: and (d) applying makeup on the face of the subject person in accordance with the makeup procedure derived in the step (c).
Another makeup simulation method embodying the present invention and another makeup method using this simulation method, particularly, a makeup simulation method for applying makeup based on dark and bright states on a face and a makeup method using this simulation method, include the following steps: (e) determining a desirable face and performing image processing on an image of a face of a subject person based on the desirable face, thereby forming images of imagined post-makeup faces; (f) displaying the imagined post-makeup faces, prepared in the step (e), on a monitor screen or the like and selectively determining an ideal post-makeup face desired by the subject person from the displayed imagined post-makeup faces; (g) performing image processing to segment an image of the ideal post-makeup face, determined in the step (f), to a plurality of areas in regard to bright areas and dark areas; (h) acquiring color data for the segmented areas obtained in the step (g); (i) determining foundations to be applied on individual areas based on the color data for those areas, acquired in the step (h), and a previously set foundation program; and (j) separately applying foundations, determined in the step (i), on the face of the subject person in association with the individual areas.
The makeup method based on dark and bright states may be modified in such a way as to be able to print the image of the ideal post-makeup face, segmented into a plurality of areas in regard to bright areas and dark areas in the step (g), in an actual size of the face of the subject person, to cut out individual areas from the printed image to prepare a pattern mask for foundation, and to execute separate application of foundations in the step (k) by using this pattern mask. This design can allow different foundations to be separately applied on the associated areas easily and accurately. That is, separate application of different foundations associated with an ideal post-makeup face can be carried out easily and accurately.
The makeup method which is based on the highlight theory may be modified in such a manner as to be able to project the image of the ideal post-makeup face, segmented into a plurality of areas in regard to bright areas and dark areas in the step (g), in an actual size on the face of the subject person, and to execute separate application of foundations in the step (j) based on this projected image. This modification can also permit different foundations to be separately applied on the associated areas easily and accurately.
Another makeup method according to the present invention which is based on dark and bright states includes the following steps: (A) determining a desirable face and performing image processing on an image of a face of a subject person based on the desirable face, thereby forming images of imagined post-makeup faces; (B) displaying the imagined post-makeup faces, prepared in the step (A), on a monitor screen or the like and selectively determining from the displayed imagined post-makeup faces an ideal post-makeup face the subject person desires; (C) performing image processing to segment an image of the ideal post-makeup face, determined in the step (B), to a plurality of areas in regard to bright areas and dark areas; (D) acquiring color data for the segmented areas obtained in the step (C); (E) determining foundations to be applied on individual areas based on the color data for those areas, acquired in the step (D), and a previously set foundation program; (F) printing foundations, determined in the step (E) in association with the plurality of areas segmented in regard to bright areas and dark areas in the image processing in the step (C), on a thin base material to prepare a transfer film; and (G) transferring foundations, printed on the transfer film prepared in the step (F), on the face of the subject person to thereby apply foundations, associated with the individual areas, on the face of the subject person.
A further makeup method according to the present invention which is based on dark and bright states includes the following steps: (H) determining a desirable face and performing image processing on an image of a face of a subject person based on the desirable face, thereby forming images of imagined post-makeup faces; (I) displaying the imagined post-makeup faces, prepared in the step (H), on a monitor screen or the like and selectively determining from the displayed imagined post-makeup faces an ideal post-makeup face the subject person desires; (K) printing an image of the ideal post-makeup face in the bright and dark states, determined in the step (I), on a thin base material using foundations to prepare a foundation transfer film; and (L) transferring foundations, printed on the foundation transfer film prepared in the step (K), on the face of the subject person to thereby apply the foundations on the face of the subject person.
Those methods can repeatedly realize makeup for an ideal post-makeup face by carrying out simple procedures, such as transferring foundations from a transfer film accurately corresponding to foundations for the ideal post-makeup face and properly shading off the transferred foundations at the boundaries between the individual areas. Thus, desired makeup can be carried out more easily and in a shorter time.
A makeup support apparatus according to the present invention for use in the foregoing makeup simulation methods and makeup methods comprises image storage means for storing images of a plurality of model faces; imaging and input means for imaging and inputting a face of a subject person; a monitor for displaying a necessary image; and processing means capable of performing a process of causing an image of the face of the subject person to approach a desirable face already determined to thereby form images of imagined post-makeup faces, a process of displaying the imagined post-makeup faces on a screen of the monitor, and a process of deriving a makeup procedure from a determined ideal post-makeup face.
It is preferable that the processing means in the makeup support apparatus may further perform a process of segmenting the determined ideal post-makeup face to a plurality of areas in regard to bright areas and dark areas, a process of acquiring color data for the segmented areas, and a process of determining foundations to be applied on individual areas based on the acquired color data for those areas.